LLM EXPLORER 63,892 MODELS INDEXED

DeepSeek V4 Pro Qwen3.5 4B by Jackrong

By Jackrong · 3476 downloads
HF Repo ↗ Copy URL

DeepSeek V4 Pro Qwen3.5 4B is an open-source language model by Jackrong. Features: 4b LLM, VRAM: 9.3GB, License: apache-2.0, LLM Explorer Score: 0.3.

  Agent Base model:finetune:unsloth/qw...   Base model:unsloth/qwen3.5-4b   Conversational Dataset:jackrong/deepseek-v4-d... Dataset:jackrong/deepseek-v4-p...   Deepseek   Distillation   En   Endpoints compatible   Gguf   Image-text-to-text   Math   Mtp   Qwen3 5   Reasoning   Region:us   Safetensors   Sft   Sharded   Stem   Tensorflow   Unsloth   Zh
Model Card on HF 🤗: https://huggingface.co/Jackrong/DeepSeek-V4-Pro-Qwen3.5-4B 

DeepSeek V4 Pro Qwen3.5 4B Parameters and Internals

LLM NameDeepSeek V4 Pro Qwen3.5 4B
Repository 🤗https://huggingface.co/Jackrong/DeepSeek-V4-Pro-Qwen3.5-4B 
Model Name/workspace/renji-training/Jackrong/Qwen3.5-4B
Base Model(s)  unsloth/Qwen3.5-4B   unsloth/Qwen3.5-4B
Model Size4b
Required VRAM9.3 GB
Updated2026-08-09
MaintainerJackrong
Model Typeqwen3_5
Model Files  5.3 GB: 1-of-2   4.0 GB: 2-of-2
Supported Languagesen zh
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Tokenizer ClassQwen2Tokenizer
Padding Token<|vision_pad|>
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to DeepSeek V4 Pro Qwen3.5 4B

Best Alternatives
Context / RAM
Downloads
Likes
AgentHorse 4B256K / 8.4 GB7607
LightOnOCR 3 4B0K / 9.1 GB239571
Qwen3.5 4B0K / 9.3 GB6575502801
NuExtract30K / 0.2 GB441570324
FrogNano 4B 26090K / 9.3 GB733113
Agents A1 4B0K / 9.1 GB17757872
Intern Decision 4B0K / 0.1 GB33848
Tev1 4B Experimental0K / 9.3 GB153916
H2o Lightning 4B0K / 9.1 GB26116
DeskForge Qwen3.5 4B0K / 9.1 GB13195
Note: green Score (e.g. "73.2") means that the model is better than Jackrong/DeepSeek-V4-Pro-Qwen3.5-4B.